Heavy metal fluoride glasses have very low intrinsic loss of light in the near-infrared and mid-infrared range. Therefore, according to Tran et al., Journal of Lightwave Technology, Vol. LT-2, pp. 556-586 (1984) they are attractive for high-energy infrared laser transmissive optics and for ultralong-length fiber optic links. However, various impurities, such as transition metal ions, can produce significant and undesirable absorption in these glasses even when present in very low levels, typically in the part-per-billion range. It is therefore essential to remove ionic contaminants from the ingredients of these glasses and, in particular, from the water-soluble zirconyl compounds and the other compounds used as starting materials in the production of the glasses.
Typical fluorozirconate glasses for optical applications consist of the fluorides of zirconium, barium, lanthanum, aluminum and sodium. Hafnium fluoride is substituted for zirconium fluoride in the cladding region of the fiber. Starting materials containing Zr, Ba, La, Al, Na and Hf therefore require purification to remove contaminants, in particular transition metal ions. However, fluorozirconate glass fibers can also be prepared using other formulations, containing fluorides of metals such as Gd, Li, Pb, Th, Ca, Y instead of, or in addition to, one or more of the components listed above. Zirconium is usually of major concern because of its high levels in common heavy metal fluoride glass formulations. The corresponding starting materials are water-soluble zirconyl compounds, which can be purified in solution, and then subjected to precipitation of zirconium as the hydroxide or oxide. The most important transition metal contaminants in fluoride glasses and in their precursors include Co, Fe, Ni and Cu.
Other uses of compounds of various metals also require a high degree of purity with respect to contaminants such as Co, Fe, Ni and Cu. High-purity lanthanum oxide for instance, is used in producing optical glass with high refracting index; other rare earths are used in solid-state devices. Solid-state lasers provide one application where the presence of certain impurities at low levels can have highly detrimental effects. In these cases and others the presence of small amounts of the contaminants mentioned above can have major undesirable effects on optical, electronic and magnetic properties.
In general, the transition metals are often divided into several subgroups. For instance, according to the book "Inorganic Chemistry of the Transition Elements", Volume 1, edited by B.F.G. Johnson, The Chemical Society, London, 1972, these elements can be divided both by row, distinguishing early transition metals such as those belonging to Groups IIIa and IVa (the scandium and titanium groups, respectively) of the Periodic Table from late transition metals, and according to period, distinguishing the first transition series from the second and third series, which include less reactive, more noble metals. For present purposes, the contaminants to be removed using the methods of this invention are late transition metals, defined as those belonging to Groups Va, VIa, VIIa, VIII, Ib and IIb of the Periodic Table. (Group nomenclature is according to F.A. Cotton and G. Wilkinson, Advanced Inorganic Chemistry, 4th ed., Wiley-Interscience, New York, N.Y., (1980)). Many of these elements, in particular those in the first transition series, have oxidation states in which they have strong optical absorption in the near-uv, visible, near-ir and mid-ir ranges. Many of these oxidation states are also paramagnetic.
Many methods have been used to separate late transition metal contaminants from compounds of other metals, including chemical precipitation or co-precipitation, fractional crystallization, distillation and sublimation. The most widely used methods in large-scale, high-efficiency purification and removal of late transition metal impurities include solvent extraction and ion exchange.
According to J. Korkisch and A. Farag, "Analytical Chemistry of Zirconium. II. Enrichment of Zirconium as Negatively Charged Sulphate Complex on the Strongly Basic Ion Exchanger Amberlite IRA-400 and Its Separation from Thorium, Titanium, Iron, Aluminum and Many Other Elements", Zeitschrift fuer analytische Chemie, 144, 81-88, (1959), the fact that zirconium is strongly retained on Amberlite IRA-400 or Dowex 1 or 2 anion exchange resins from a 0.1 N sulfuric acid solution allows zirconium to be separated from many elements, including copper, trivalent iron, cobalt and nickel. Zirconium is subsequently eluted from the bed with 4 N hydrochloric acid. According to F.W.E. Strelow, "Separation of Zirconium from Titanium, Ferric Iron, Aluminum, and Other Cations by Cation Exchange Chromatography", Analytical Chemistry, 31, 1974-1977, (1959), the sorption of zirconium on a BioRod AG50W, X8 cation exchange resin from a 2 N hydrochloric acid solution has been used for its separation from many elements, including divalent and trivalent iron, copper, nickel and cobalt. Zirconium is subsequently eluted with 5 N hydrochloric acid.
These and similar ion exchange methods generally have several disadvantages. Zirconium is sorbed on the ion exchange bed in preference to the impurities, and this limits both the efficiency and the capacity of the column and requires elution to be carried out whenever a small volume of zirconium solution has passed through the medium, provided the solution has a moderately high concentration of Zr. Since the capacities of cation exchange resins are usually less than 5 meq/mL and those of anion exchange resins less than 2 meq/mL, it follows that no more than 1.25 volumes of a 1 M Zr solution (91 g/L or 5 meq/L of Zr) can be sorbed on a volume of cation exchange resin and no more than 0.5 volume of the same solution on a volume of anion exchange resin. Furthermore, the eluted Zr solutions in both cases are highly acidic and they require a large amount of base to precipitate zirconium hydroxide or hydrous oxide, resulting in a large increase in volume and risk of re-contamination. It is therefore very desirable to sorb late transition metal impurities preferentially on the ion exchange or sorption medium from the zirconium solution, and this requires the medium to have higher affinity for these impurities than for zirconium. In the case of conventional cation exchange resins the ions of Zr and of similar elements, e.g. Hf, La and Al are generally sorbed on the resin in preference to ions of late transition metals such as Fe, Co, Ni and Cu because of the higher charge density (charge-to-size ratio) of the former ions. In the case of conventional anion exchangers, the extent of sorption of various metals depends on the stabilities and charge densities of the respective anionic complexes. In the case of high chloride concentrations, for instance, Co, Cu and Ni adsorb on anion exchange resin to a lesser extent than Zr, Hf and Ga, according to R.M. Diamond and D.C. Whitney, "Resin Selectivity in Dilute to Concentrated Aqueous Solutions", Chapter 8, in the book "Ion Exchange", Vol. 1, by J.A. Marinsky, Marcel Dekker, New York, N.Y., 1966, pp. 277-351. According to the same authors, Co, Ni, and Cu are not adsorbed on anion exchangers from nitrate media. Accordingly, conventional anion exchange resins are not expected to be useful in removing dissolved late transition metals from solutions containing large amounts of other dissolved metals such as Zr, Hf and Ga.
Several of the most important species which require purification for optical, electronic and other applications, such as Zr, Hf, or, to a lesser extent, La and Bi, as well as other multivalent metals, are stable in solution only at very low pH values, typically below pH 2-3. This hinders their purification by means of cation exchange resins and even chelating ion exchange resins such as BioRad Laboratories Chelex 100 which show a drastic decrease in capacity as the pH decreases from about 4 to below 2.
According to Klein et al., Optical Engineering, Vol. 24, pp. 516-517 (1985), standard ion exchange methods can be used to remove iron, cobalt, nickel and copper from zirconyl ion solution. However, column capacities are not given, and they can be expected to be very low based on the previous references, which show zirconium ions to have higher affinity to both cation exchange and anion exchange resins. The residual concentrations after ions exchange in the cases of Fe, Co and Ni are 0.38, 0.36 and 1.15 ppm, respectively, or 5%, 5% and 16%, respectively, of their initial levels in the influent.
It is therefore very desirable to have an ion exchange or sorption medium with high selectivity for late transition metal impurities in the presence of relatively high concentration of ions with high charge density. Such selective sorption also permits pre-concentration of trace impurities on the solid medium for analytical purposes.
Various organic compounds are known to form complexes with late transition metal ions which can be used in solvent extraction procedures of separating these ions from aqueous solutions.
Heteropolycyclic compounds based on the bipyridine or bipyridyl structure: ##STR1## or NC.sub.6 H.sub.4 -C.sub.6 H.sub.4 N structure are known to form colored complexes with certain late transition metal ions, in particular ferrous and cuprous ions. In general, a heteropolycyclic compound can be defined as a compound having at least two rings which are not fused directly together, with one or more heteroatoms, in particular nitrogen atoms, in the structure of at least one of the rings. According to the books "The Iron Reagents.revreaction. (1960) and "The Copper Reagents.revreaction. (2nd ed., 1972), both by H. Diehl and G.F. Smith, published by The G. Frederick Smith Chemical Company, Columbus, Ohio, many compounds which include the NC.sub.6 H.sub.4 -C.sub.6 H.sub.4 N structure form complexes with the ions mentioned above. These include 1,10-phenanthroline: ##STR2## and its derivatives such as 4,7-diphenyl-1,10-phenanthroline (bathophenanthroline), 2,9-dimethyl-1,10-phenanthroline (neocuproine), 2,9-dimethyl-4,7-diphenyl-1,10-phenanthroline (bathocuproine) and 2,2'-biquinoline or 2,2,'-biquinolyl (cuproine). Other heteropolycyclic compounds which form similar complexes have structures which are not directly based on the bipyridine or phenanthroline structure but all of them contain two non-fused rings with at least one ring containing a nitrogen atom. Such compounds include 2,4,6-tripyridyl-s-triazine (TPTZ) and syn-phenyl-2-pyridyl ketoxime.
According to Hach, U.S. Pat. No. 3,095,382, iron in water is determined colorimetrically with a mixture of about 40% Na.sub.2 S.sub.2 O.sub.4, 60% NaHSO.sub.3 or Na.sub.2 S.sub.2 O.sub.5, and 4% of a heterocyclic compound, such as 1,10-phenanthroline. About 0.1 g of the mixture added to 25 mL of water dissolves oxides, reduces the iron, and buffers the sample at pH 6.5 where the color is developed. The composition of the mixture can be varied over wide limits.
According to the books "The Iron Reagents.revreaction. and "The Copper Reagents.revreaction. by H. Diehl and G.F. Smith cited above, complexes of heteropolycyclic compounds with late transition metal ions can be extracted into solvents such as isoamyl alcohol or 3-methylbutan-1-ol, 1-hexagonal and nitrobenzene. However, the use of a solid sorbent or ion exchange medium offers many advantages over solvent extraction in the removal of trace impurities. In particular, the use of packed columns or beds provides a large number of equilibration and separation stages, known as theoretical plates, over a short length. The problems of liquid phase miscibility, of impurity concentration at interfaces, of contamination of the product stream with the solvent and of the presence of hazardous solvent vapors are avoided.
Furthermore, according to the books "The Iron Reagents.revreaction. and "The Copper Reagents.revreaction. by H. Diehl and G.F. Smith cited above, the heteropolycyclic compounds which form complexes with late transition metal ions are usually specific to only one or a few of these ions. For instance, compounds bearing substituent groups on the carbon atoms adjacent to the ring nitrogen atoms of the bipyridine or phenanthroline structure form colored complexes with copper, but not with iron. Accordingly, solvent extraction techniques have traditionally been based on the use of different complexing compounds for different ions, e.g. 1,10-phenanthroline, bathophenanthroline, TPTZ and phenyl-z-pyridyl ketoxime for iron and cuproine, neocuproine and bathocuproine for copper.
Various organic chelating agents, including heteropolycyclic compounds such as 1,10-phenanthroline (o-phenanthroline) and neocuproine, as well as other chelating agents such as 8-hydroxyquinoline (oxine), dithizone and salicylaldoxime, can also be coupled to insoluble inorganic carriers such as silica, silica gels, glass, porous glass, bentonite, hydroxyapatite, alumina, and nickel oxide by means of an intermediate silane coupling agent, according to Shucker et al., U.S. Pat. No. 3,886,080. According to this patent, the chelating agent is bonded to the silane coupling agent by means of an azo linkage. The immobilized chelating agents can be used to remove trace metals such as Fe, Cu, Mo, Zr, V, W and Ti, from solution. However, silanization is a complicated and expensive technique. Used to remove trace metals from raw materials for fluoride glass production or other applications it can lead to contamination with undesirable traces of Si. The expected loading of the support with the chelating agent is low, and, in the Example listing efficiencies for the removal of various metals, using a porous glass-oxine material, zirconium is observed to be effectively sorbed, indicating that such combinations are not suitable for selective separation of late transition metal impurities from zirconium solutions. According to U.S. Pat. Nos. 4,421,654 and 4,448,694 by Plueddenann transition metal ions can be removed from solution using a method which comprises contacting a solution containing these ions with an inorganic solid substrate, such as silica gel, treated with a silylating agent which a polyamine derivative. This process does not make use of substrates based on the bipyridyl or phenanthroline structure.
British Pat. No. 1,355,535 describes a method for extracting a metal from a solution containing an ion of this metal which comprises contacting the solution with an adsorbent which is peat or brown coal or a brown coal char in the presence of ammonia capable of forming a stable complex with the metal. The ammonia derivative is an aliphatic amine, a hydroxy-aliphatic amine or a nitrogen-containing heterocycle such as pyridine. There is no mention of heteropolycyclic compounds. This method is used to recover metal values from solution but the patent does not address selective removal or separation of various metals such as the removal of late transition metal ions from solutions of Zr, Hf, Al, La or other multivalent metal ions. Furthermore, a major difference between this patent and the present invention is that according to the patent, the ammonia derivative is added to the metal solution rather than used to pre-treat the carbonaceous support. This process is much less suitable for decontamination of nuclear plant effluents or other streams than the use of a pre-treated support according to the present invention.
U.S. Pat. No. 4,222,892 by Motojima et al. describes a process for preparing charcoal impregnated with oxine (8-hydroxyquinoline) which comprises contacting solid oxine with the charcoal until the charcoal becomes impregnated with the oxine. This Patent mentions the use of supported oxine to remove heavy metals from solution but does not address selectivity in such removal and does not mention supported reagents other than oxine.
U.S. Pat. No. 4,659,512 by Macedo et al. describes a process for removing metal species from solution using porous silicate glass, silica gel or charcoal containing an alkylene amine. Heteropolycyclic compounds are not used.
U.S. Pat. No. 3,917,453 by Milligan et al. mentions 1,10-phenanthroline or 4,7-diphenyl-1,10-phenanthroline coated on silica gel to assist in obtaining a permanent record of the determination of glucose by means of a ferric salt. This patent does not address removal of metal late transition ions from solution or their separation from other ions.
In addition to removing dissolved late transition metals from solution, it is often desired to separate them from each other, for instance in the preparation of catalysts based on individual noble metals of the last two rows of Groups VIIa and VIII of the Periodic Table.
Furthermore, in addition to the problem of removing dissolved late transition metals from solutions, it is often necessary to immobilize and concentrate such metals from various media for purposes such as management of radioactive or toxic metal species (e.g. Co-60, Co-58 or Fe-59), recovery of precious metals (e.g. Pt and Pd), or accurate analysis of trace metals.
It is an object of this invention to provide a simple and convenient process of removing late transition metal species from liquid streams.
Another object of this invention is to make sorption media which will selectively remove dissolved late transition metal species such as Fe, Co, Ni and Cu from liquid streams in the presence of large excess amounts of other dissolved metal species such as Zr, Hf, La, Al, Ga, In, Sc, Y and Ba.
Yet another object of this invention is to provide a method for removing dissolved late transition metals from acidic solutions.
It is also an object of this invention to provide a process for producing highly pure starting materials for the preparation of highly transparent optical glasses and fibers, in particular fluorozirconate glasses and fibers.
It is also an object of this invention to provide a process for the selective recovery of particular dissolved late transition metals from liquids that contain more than one dissolved transition metal.
It is a further object of this invention to provide sorption or ion exchange material in order to concentrate low levels of dissolved late transition metals for purposes of radioactive or toxic species management, valuable metal recovery, and trace analysis.